In vivo endotoxin exposure increases neutrophil counts ... - CiteSeerX

1 downloads 0 Views 1MB Size Report
sheep showed primed superoxide anion release compared with controls when ...... Jacob,. H.S.. (1990). Endothelial cell platelet-activating factor primes neu-.
In vivo endotoxin without

maintenance

short-term

A. Carey,

Department Medical

to

of

of sheep

Kurt

H. Albertine,

Physiology

College,

increases

of primed

cultures

Lisa

Abstract: lished responses endotoxin

exposure

and

Thomas

Division

Jefferson

of

superoxide

bone

and

Marlys

Pulmonary

University,

neutrophil anion

release

H. Gee

Medicine

and

Critical

Care,

Department

of

superoxide

respond

compared

with

controls

when

stimulated with phorbol myristate acetate. Priming was no longer present in neutrophils tested after 2 days in cubture. The neutrophil cell count, however, was higher after 2 days in culture than in controls. Increased neutrophil production/viability was reproduced in naive bone marrow cultures by replacing part of the normal serum medium with plasma taken from sheep several hours after endotoxin infusion. These data suggest that a dissociation exists between tion/viability

priming and increased in response to endotoxin.

56:

1994.

145-150;

Key bility

neutrophil

J.

endotoxin . sheep #{149} bone marrow . superoxide anion release

Words: . priming

. neutrophil

.

to activating

MATERIALS Culture

of of

medium was prepared as described by Greenberg et al. Briefly, the following were added to 800 ml of McCoy’s 5A media: 6 ml of7.5% NaHCO3, 20 ml ofsodium pyruvate (100 mM), 4 ml (lOOx) of minimal essential medium (MEM) vitamin mixture, 8 ml of Eagle’s MEM amino acid, 4 ml (lOOx) of Eagle’s MEM nonessential ml of 21 mg/ml L-serine, 1.6 ml of 10 mg/ml 5 ml of glutamine, 250 mg of streptomycin, penicillin, 142 ml ofsterile fetal bovine serum ml of sterile horse serum (HS) to achieve a 12.5% HS medium. All chemicals were

via-

Gibco endotoxin

physiological

in

experimental

responses,

functional response of inflammatory phages and neutrophils. We and that peripheral blood neutrophils

animals including

changes

causes in

BRL,

further

characterizing

the

cells such as macroothers have demonstrated isolated hours after en-

duration

of

this

The

Collection Bone sterile

Island,

NY.

The

medium

was

filter-

use.

relative maturity of the cells, neutrophils, that are capable of which cause priming is unknown. of the duration of priming in bone short-term culture method for sheep purpose was to determine the dura-

of

1 ml 9 ml

was drawn in anesthetized

heparin

t\t)t)rt’viat

was

injected

inns:

C : P. (-Utst

st-ruin;

v1EN1, Rrpn Collegt-,

IN

acetate; nt

request

Ihis

New Received

IN

crest a lung

needle,

and

during lymph

the

Jeflerson

Health work

F, t ii taut Hall

current

Sciences was

Rooni

431.

address:

Center, in part

presented

I)t’pt

. (‘t’.

I-IS,

actor:

optical

FBS. Ic(N4-CSF,

actor; (I(’nsitv;

horse

seroiris;

PN’IA_

photbol

factor.

necrosis

H

Alutnni

ing

()l),

other

for the contents

1)liStthi

I ). PSt(’t5(ItXit5

colonv-stitnulattng

tiiediutn;

s: NI arlvs

19107. H. Albertine’s

Utah

l)l1sttsa:

granulocvte

(‘ss(’ntial

Kurt

Journal

the

(olonv-stiiosulat

tninitnal

ti)VtIstate

in

the right iliac sheep to prepare

through

ml

S1-I

(;-(

granulocyt(’-rnacrophag(’

of

from

was retained in the syringe as an anticoagulant of bone marrow drawn into the syringe. The

toil bovine

PA

of bone marrow

and place vascular lines [5]. The bone marrow was into a 13-gauge Jamshidi needle with an adaptor to 10-mb syringes were attached. Each of the six to eightused contained 2 ml of sterile heparin. The first 1

syringes ml

and culture

marrow surgery

fistula drawn which

neutrophil

functional change. Neutrophils are short-lived cells with a circulating half-life of6 to 8 h [3]; therefore, it is evident that blood samples taken 24 h after endotoxin infusion contain leukocytes that were released from bone marrow during or after the endotoxemia. from stem cells to mature responding to the signals To address the question marrow, we developed a bone marrow cells. Our

Grand

before

amino acids, 0.4 L-asparagine, 250,000 units of (FBS), and 142 12.5% FBS and purchased from

a the

dotoxin infusion show primed superoxide anion release when stimulated in vitro [1, 2]. Priming is also evident in sheep bone marrow neutrophils isolated 24 h after a bolus endotoxin infusion [1]. Sustained priming of superoxide anion release, particularly in immature bone marrow cells, suggests that this functional modification of neutrophils may be part of a long-term inflammatory response to endotoxin. Given the persistence of priming in both peripheral blood and bone marrow-derived neutrophils, we became interested in

AND METHODS

media

sterilized Infusion

to

[4].

[4].

INTRODUCTION

myriad

agents

access to large numbers that are physiologically content and ability

The

producBiol.

Leukoc.

Jefferson

tion of priming by culturing bone marrow from sheep after exposure to endotoxin. We modeled our system after that described by Greenberg et al. [4], which is a long-term liquid culture method. This method is particularly well suited for

our studies, because it allows ready of bone marrow-derived neutrophils normal in terms of their granular

release

Medicine,

Pennsylvania

from surgical control animals and from sheep given a 12-h infusion of endotoxin. In these bone marrow aspirates, neutrophils from endotoxin-treated sheep showed primed

anion

in

marrow

Philadelphia,

A short-term bone marrow culture was estabdetermine the functional and proliferative of neutrophils sampled from sheep exposed to in vivo. Iliac crest bone marrow was drawn

counts

.

of Physiology, 1020

locust

I)epartment

Salt

lake

at the

.Jr’tIrson Street,

of Pediatrics,

City,

Utah,

Experimental

Medical

Philadelphia, University

84132-1001. Biology

93

meeting

Orleans. Novetsiber

of Leukocyte

16,

1993;

Biology

accepir(1

Volume

Ntarch

56,

31.

August

199-1.

1994

145

of the

syringes

centrifuge calcium-free centrifuged

were

transferred

tubes each balanced for 15 mm

to two

sterile,

containing 5 ml salt solution (Gibco). at 900g in a Beckman

capped

of

50-mb

sterile Hanks’ The tubes were model TJ-8 cen-

trifuge.

The buffy coat was aspirated, subjected to hypotonic to remove the red blood cells, and then resuspended sterile, 2 x Hanks’ balanced salt solution (Gibco) to the

nucleated

cells

to isotonicity.

The

cells

were

lysis in return

centrifuged

at 2500g in a Sorvall refrigerated centrifuge. The washed twice with McCoy’s modified 5A medium trifuged again at 2500g. The cells were resuspended of medium and the nucleated cells were counted

pellet was and cenin 10 ml using a

Peripheral

blood cultures

To determine under these blood-derived

the viability of peripheral blood culture conditions, we cultured white blood cells. Forty-five

sheep blood was obtained

was drawn as described

McCoy’s row cells.

5A medium The cultures

counted,

and

marrow

5 ml above,

counts

exposure

drawn

from

sheep during sterile surgery periments. These animals

of

of heparin. The buffy coat and the cells were plated in

as previously described were harvested at

differential

In vivo endotoxin Bone

in

neutrophils peripheral milliliters

for bone 2 days, the

marcells

obtained.

protocol the

right

iliac

crest

of

normal

served as controls for these exwere allowed to recover for 4 to 5

hemacytometer. Some nucleated red blood cells survived hypotonic lysis and were present in the culture. The cells were seeded in a 20-mb volume in a T-75 cell culture flask (Corning Plastics, New York, NY) at a concentration of 4000 cebls/il. A portion of the cells was saved for a baseline differential cell count using Wright-Giemsa-stained slides

days before the endotoxin consisted of a low-dose,

and a superoxide anion remainder of the cells tamed at 33#{176}Cin room 2 days.

prepared for a sterile necropsy. At this time, bone marrow was drawn from the left iliac crest and cultured as described above. When a bone marrow aspirate was not successfully obtained at the time of surgery, a bone marrow aspirate was

Superoxide

anion

release assay as described below. The were placed in an incubator mainair supplemented with 3% CO2 for

release

assay

Bone marrow cells, freshly aspirated or cultured, were trifuged at 2SOOg and resuspended in sV-2 ethanesulfonic acid (HEPES) buffer containing (in mM): 140 NaCl, HEPES, 10 KC1, 0.1 CaCl2, 0.2 MgCl2, 11.0 NaHCO3, glucose, and 14 iM albumin (Sigma, St. Louis, MO), 7.4. Because the bone marrow contained heterogeneous types, we diluted the cells to a concentration of 250,000 what we termed superoxide anion-releasing cells

cen20 5.0 pH cell of (eo-

sinophils, bands, and segmented neutrophils/50 jl) and plated them in Nunc high-binding microtiter wells. The cells were allowed to settle for 10 mm. The wells were stimulated with 50-l test solutions containing 10” through 10 M phorbol myristate acetate (PMA; Sigma) and 0.32 mM ferricytochrome c (type III from horse heart, Sigma). Triplicate wells received each treatment. Three wells received ferricytochrome c and HEPES buffer to measure basal release of superoxide anion. There were also three assay blank wells that contained ferricytochrome c, HEPES buffer, and 700 U superoxide dismutase (bovine erythrocytes, Calbiochem) to confirm that by superoxide

the

ferricytochrome dismutase.

c reduction

was

inhibitable

Changes in optical density (OD) were measured intermittently on a microplate reader (Dynatech MR 600) using a test wavelength of 550 nm and a reference wavelength of 670 nm. Superoxide anion release was calculated using the conversion proposed by Pick and Mizeb [6]: nmol O2 = [(test OD

-

reference

Collection After bator. medium

OD)

x

100]

6.3.

at 2 days

2 days,

the

culture

The adherent using a cell

flasks cells scraper

were

removed

from

were gently scraped (Gibco). A cell count

the into was

incuthe ob-

tamed and percent viability was determined using trypan blue exclusion. Cell counts were reported for viable cells only. Smears were made for total cell differential counts. The cells were centrifuged at 2500g for 5 mm, and the cell pellet was resuspended in HEPES. The cells were plated in microtiter wells as described above and a superoxide anion release assay was completed.

146

Journal

of Leukocyte

Biology

Volume

56,

August

1994

infusion was 12-h infusion

(Sigma) at 10 ng/kg-min, as described After a 12-h recovery period, the sheep

begun. of E.

The infusion coli endotoxin

by Sloane et was anesthetized

al.

[7]. and

drawn from a control animal. Because the sheep used for our studies are raised domestically, they are subject to environmental stresses that change seasonally. Bone marrow characteristics may also vary with the age of the animal. To avoid issues of age and seasonal variation, we chose age-matched and seasonally matched sheep to serve as controls when a bone marrow aspirate was not successfully obtained at the time of surgery.

Replacement

studies

To determine whether the effects of in vivo endotoxin infusion on these bone marrow cultures are attributable to a plasma-borne factor, we collected heparinized blood from two sheep 7 h after the end of the 12-h endotoxin infusion, at a time point characterized by peripheral blood leukocytosis [7]. The blood was centrifuged for 15 mm at 900g in a Beckman model TJ-8 centrifuge to separate the blood components. Plasma was then removed and frozen. The plasma samples from both sheep were pooled, combined with an equal volume of FBS, filter sterilized, and frozen in 5-mi aliquots until needed. Control plasma was collected from two sheep, after surgery but before experimental intervention, and was likewise processed and combined with an equal volume of FBS. Bone marrow was then drawn from a surgical control animal and subjected to the normal separative procedures. The nucleated cells were suspended in serumbess medium and were plated in three T-75 flasks containing the following combinations with 12.5% FBS: 12.5% horse serum (HS), the normal culture medium; 12.5% postendotoxin plasma (END); and 12.5% control plasma (CP). An initial cell superoxide anion release assay was obtained, as were superoxide anion assays on the contents of the three flasks after 2 days in culture. Total cell count, differential, and percent viability

were

also

obtained

In vitro endotoxin

as

described

above.

studies

To determine whether incubation with endotoxin in vitro would change neutrophil cell function and viability, we added E. coli endotoxin in I ml of saline to naive cultures to achieve final endotoxin concentrations of 10 ng/250,000 nucleated cells (160 ng/ml) and 0.1 ng/250,000 nucleated

cells (1.6 ng/ml). to be associated while the lower plate

The higher concentration has been shown with neutrophil respiratory burst activation, dose is associated with priming [8]. A third

received

differential, tamed for

saline the

to serve

as a vehicle

and superoxide three treatment

anion groups

control.

release after

Cell

assays 2 days

TABLE

Properties

of

were obin culture.

Initial studies were conducted to characterize the culture systern. We were able to support granulocyte production for 14 days using a biweekly medium replacement schedule. This suggests that the culture conditions were compatible with stem cell survival and colony formation. Table 1 summarizes the data gathered from surgical control bone marrow cubtures harvested after 2 days. Note that neutrophil counts This where

is in conthe neu-

trophil counts dropped 71% ± 8% over the 2-day period. The ability to release superoxide anion in response to PMA was maintained over the 2 days in culture, suggesting that the neutrophils which were present at 2 days were healthy and were not down-regulated or primed by the culture conditions. PMA-stimulated superoxide anion release assays were performed on unfractionated bone marrow samples that were not subjected to any separative procedures to isolate the neutrophibs. marrow

We samples

performed some using discontinuous

separations of Percoll/plasma

the

bone gra-

dients as previously described [9] before determining the superoxide release. Release from these isolated granubocytes was significantly higher than the release from unfractionated bone marrow: 4.05 ± 0.6 nmol O2 versus 2.25 ± 0.3 nmol O2 (n = 10, P < .05, t-test, two-tailed). An abbreviated version ofour superoxide anion assay requires 3-4 million neutrophils. Given that these cultures normally yield approximately i8 million neutrophils after 2 days, we were not always able to obtain enough cells in any one layer of the gradient to perform the superoxide anion assay. For this reason, we used unfractionated bone marrow in the assays, even though we found higher release from fractionated cells.

-0---

-

Marrow

Short-Term

Cultures

Value

Fresh

marrow’

2-day

bone

marrow

4000 39

(P

unpaired

Bone

3

±

133 0.34

±

from 1-test,

marrow

1487

±

127

61

±

2’

885 2.27

±

80’

±

0.35

granulocytes)

difference

.05,